In C, accessing an array index is faster or accessing by pointer is faster? By faster I mean, which one would take less clock cycle. The array is not an constant array.
At the lowest level, these operations mostly tend to compile to the same thing. If you're really interested, you should get your C compiler to generate assembly output (such as with gcc -S) so you can check, especially since it depends, at a bare minimum, on:
You'll find that, even if there was a difference (which is doubtful), that level of micro-optimisation is mostly not worth the effort you put into it. You're better off doing macro-optimisations such as improved algorithms since that's the sort of thing that offers more return on investment.
In these sorts of situations, where the effect is likely to be minimal, I always optimise for readability.